Posted on | November 5, 2012 | 24 Comments
Prudence Paine had asked me to submit an Electoral College prediction map and, while Smitty played the game, I chose to abstain. It’s just not what I do. I once won an office pool by correctly guessing the length of a Bill Clinton State of the Union speech, but I suck at election predictions.
Uh, 401 Electoral College votes? C’mon. There are plausible scenarios in which Romney could clear the 300-ECV threshold, and that’s about as high as I’d be willing to go. Just imagine that you had $100 riding on the outcome, and the outrageous optimism immediately chills. Give me Nate Silver’s 7-to-1 odds against Romney, and I’d gladly bet $100, but other than that, I’ll predict nothing.
As I’ve said before, it’s better to think in terms of contingencies: “If X, then Y.” I just saw some data on early votes in Virginia that would seem to indicate Romney winning that state with 52% or more, an outcome (i.e., R+4) that no recent poll has predicted. OK, so if Romney overperforms the polling in Virginia, then he is likely to overperform in other states.
That contingency would then suggest that the systemic oversampling of Democrats that we’ve seen in so many polls has badly misrepresented the electorate, and you could probably figure on adding two or three points to Romney’s RCP average in any state. But readers didn’t send me to Ohio so I could pontificate about poll numbers. Time to leave this McDonald’s and get back on the road to the Romney rally in Columbus that starts in two hours.
Hit the freaking tip jar!